WO2011003263A1

WO2011003263A1 - Detection method and apparatus based on random access process

Info

Publication number: WO2011003263A1
Application number: PCT/CN2009/076042
Authority: WO
Inventors: 李斌
Original assignee: 中兴通讯股份有限公司
Priority date: 2009-07-06
Filing date: 2009-12-25
Publication date: 2011-01-13
Also published as: EP2439973B1; CN101944929A; EP2439973A4; JP5223036B2; CN101944929B; EP2439973A1; JP2012531870A

Abstract

A detection method based on a random access process is provided in the present invention, which includes: according to a received random access channel RACH signal, obtaining a RACH time-domain correlation value sequence; calculating a first detection threshold Th_i ¹; for a search window of each RACH time-domain correlation value sequence, when a peak value of the RACH time-domain correlation value sequence is more than the first detection threshold, recording the peak value and a timing location; in search windows, if there are one or more set(s) of continuous search windows which all have the peak value more than the first detection threshold, converting the timing location corresponding to a peak value more than a second detection threshold to be a timing adjustment quantity. The present invention also provides a detection apparatus for based on the random access process correspondingly. The present invention can reduce the false alarm probability of detection, and provide the accurate and reliable uplink timing adjustment information for UE, in order to improve the reliability of UE accessing LTE system and reduce signaling overhead caused by processing the false alarm.

Description

Detection method and detection device based on random access process

The present invention relates to the field of communications, and in particular, to a detection method and a detection apparatus based on a random access procedure. Background technique

Random Access is the access process of the User Equipment (UE) before the start and network communication. In LTE systems, random access can be divided into two types: Synchronized Random Access and asynchronous random access.

(Non- synchronized Random Access), when the UE has been uplink synchronized with the system ear, the random access procedure of the UE is called synchronous random access; when the UE has not obtained uplink synchronization with the system or lost uplink synchronization, the UE The random access procedure is called asynchronous random access. Since the UE has not obtained accurate uplink synchronization during the asynchronous random access, a main feature that the asynchronous access is different from the synchronous access is that the asynchronous access needs to estimate and adjust the uplink transmission clock of the UE, and the synchronization will be synchronized. The error is controlled within the length of the Cyclic Prefix (CP).

Generally, after starting, the UE performs downlink synchronization through a Synchronization Channel (SCH) to obtain a radio frame number, a receiving position of a subframe, and a cell ID. After that, the UE detects a broadcast channel (BCH) to obtain a broadcast channel (BCH). System information, the system information includes configuration information of a random access channel (RACH); finally, the UE performs uplink synchronization through the RACH channel to complete the process of accessing the system, and the process belongs to asynchronous random access.

3rd Generation Partnership Project (3GPP) Long Term Evolution

The (Long Term Evolution, LTE) protocol provides multiple preamble sequences for uplink random access. In the uplink synchronization transmission process, the UE obtains the radio frame and subframe position determined according to the downlink synchronization. RACH channel location, randomly selecting one of the available sequences as the preamble transmission, the base station detects it, determines the timing adjustment amount of the uplink synchronization and transmits it to the UE, and the UE transmits the data to the uplink according to the timing adjustment amount delivered. Adjustments are made to achieve time synchronization of the upstream channel.

The uplink random access preamble sequence specified in the LTE protocol is a ZC (Zadoff-Chu) sequence, and the u-th root ZC sequence is defined as:

.7mn(n+l)

x _u (n) = e ^{3 Wzc} , 0 < n < N _zc -l

Where u is the root sequence index number, Nzc is the length of the ZC sequence and Nzc is a prime number, and the value in the LTE protocol is 839 or 139.

In the LTE system, each cell allocates 64 sequences for preambles, which may be different cyclic shift sequences from the same root sequence, or different cyclic shift sequences from different root sequences. . The root sequence of ZC and its cyclic shift sequence have good autocorrelation properties, and the correlation between different sequences is small. Therefore, the base station can use the correlation characteristics of ZC sequences to perform time domain correlation detection on random access signals. Obtain the timing synchronization adjustment amount of the uplink.

The time domain correlation detection method is: multiplying and summing the received signal and the complex conjugate of each cyclic shift of the local sequence to obtain time domain correlation values of all cyclically shifted sampling points, using inverse discrete Fourier transform ( Inverse Discrete Fourier Transformation (IDFT) feature, this process can be equivalent to: After receiving the frequency domain signal and the local frequency domain sequence complex conjugate point multiplication, transform to the time domain, the mathematical form of time domain correlation detection is as follows:

Suppose the time domain of the received signal is y(m), the frequency domain is Y(k), the local sequence time domain is x(m), and the frequency domain is X(k). Then the time domain correlation function R(m) ) can be expressed as:

R(m) = £ x'(k) - Y(k) - e ^N "

m is a cyclic shift point, 'is a complex conjugate. Therefore, for a UE that uses a different cyclic shift of the same sequence as a preamble, the base station converts the received signal into a frequency domain, and performs complex conjugate point multiplication with the frequency domain sequence of the sequence. After the result is IDFT transformed into the time domain, the time domain correlation values corresponding to all cyclic shift samples are obtained. By performing peak detection on the time domain correlation value corresponding to each cyclic shift sample of the local root sequence, the preamble used by the UE can be determined, and the timing advance amount of the UE is obtained.

The performance of RACH time domain correlation detection can be characterized by missed detection probability and false alarm probability. The false alarm probability is the probability of false detection of the preamble when there is no preamble transmission; the missed detection probability is that no transmitted preamble is detected. The probability of occurrence of a code. In general, the probability of false alarms is not greater than 10 ³ .

In the detection of the above RACH time domain detection method, both the false alarm probability and the missed detection probability are determined by the threshold in the peak value detection algorithm. Generally, in the peak detection algorithm, the ratio of the correlation value of the random signal to the noise power estimation value is used as a threshold. Therefore, in the detection algorithm, the accuracy of the noise power estimation value has a large influence on the performance of the detection algorithm.

In addition, since there is a certain frequency offset in the LTE system, although the LTE protocol specifies a restricted set of preambles in the presence of a large frequency offset, the false alarm probability is reduced, but for the medium and low speed cells using the unrestricted set, The Doppler frequency offset caused by user speed will cause the false alarm to rise. Especially in the case of frequency offset and high signal-to-noise ratio, the rise of false alarm will increase the signaling overhead of the random process and increase the processing of the system. burden. Therefore, it is necessary to suppress the false alarm in this case.

In addition, since the length of the RACH sequence is prime, the Discrete Fourier Transformation (DFT) and the IDFT complexity of the prime number are high. In the general implementation, the FFT (Discrete Fourier Transform) is used. And IFFT instead of DFT and IDFT processing. In order to ensure that no information is lost, the value of the second power must be greater than the length of the ZC sequence. Therefore, oversampling is introduced in the time domain correlation operation. Although the technique can reduce the complexity of implementation and improve the time domain resolution, it also leads to various phases. The energy dispersion of the threshold points reduces the autocorrelation property of the preamble sequence, resulting in an increase in the false alarm probability.

In summary, in the existing RACH detection algorithm of the LTE system, there is a defect that the false alarm probability becomes high in the case of frequency offset and high SNR, and the false alarm probability caused by the oversampling technique becomes high, and Accurate and fast estimation of noise power, so the implementation complexity is higher. Summary of the invention

An object of the present invention is to provide a detection method and a detection device based on a random access procedure, which can reduce the false alarm probability of the detection, and provide accurate and reliable uplink timing adjustment information for the UE, so as to improve the reliability of the UE accessing the LTE system. And reduce the signaling overhead caused by handling false alarms.

According to an aspect of the present invention, a detection method based on a random access procedure is provided, the method comprising:

Performing time domain correlation processing on the received random access channel RACH signal and the local root sequence to obtain multiple RACH time domain correlation value sequences;

Determining a noise power estimation mean Γ' ^∞ or a maximum peak pr ^k in a sequence of correlation values corresponding to each RACH time domain correlation value sequence, and calculating a corresponding first detection threshold from any of the values

Thj ;

For each search window of the RACH time domain correlation value sequence, if the peak value in the RACH time domain correlation value is greater than the corresponding first detection threshold 73⁄4, the peak value and its timing position are recorded;

For each search window of the sequence of RACH time domain correlation values, when there are one or more sets of consecutive multiple search windows having peaks greater than the corresponding first detection threshold, determining peaks of the plurality of search windows Is it greater than a corresponding second detection threshold? 3⁄4, where Th _k ² , w R _p ^k _eak , is the maximum value among the peaks of the plurality of search windows, 0 ≤ ≤ l , = 1, A , L , And L represents the number of root sequences, k = , k , K , and is the number of groups of the consecutive plurality of search windows; For a peak value greater than the corresponding second detection threshold 7 \ ² , , ·, the corresponding timing position is converted into a timing adjustment amount.

Calculating the first detection threshold Th by using the noise power estimation mean ::

The first detection threshold is calculated; where, = 1, AL, and L represents the number of root sequences, and β is a noise threshold factor greater than one.

The first detection threshold is calculated by using the maximum peak value in the correlation value sequence as: the first detection threshold is calculated according to 73⁄4 = /^4; wherein, the peak threshold is in the range of 0<<1.

The determined noise power estimate mean /r' ^si includes:

According to the length of the W sequence

Ncs divides the RACH time domain correlation value sequence into N search windows;

Wcs is a preset window length parameter;

Find the peak value of the RACH time domain correlation value in each search window, and find the average value of the RACH time domain correlation value smaller than the peak value, and obtain the temporary average value of the noise in the search window, where = 1, AL, and L Represents the number of root sequences, j = l, A, N, and N is the number of search windows, 0 < « <1; the mean value of the noise average obtained in all search windows of the RACH time domain correlation value sequence, The noise power estimate mean /T' ^{se is obtained} ; where = 1, A, L, and L represents the number of root sequences.

The value is determined by a predetermined false alarm probability and a chi-square distribution characteristic.

Performing time domain correlation processing on the received RACH signal and the local root sequence to obtain multiple RACH time domain correlation value sequences includes:

Downsampling the received RACH signal;

Performing DFT transform on the downsampled data to obtain frequency domain RACH received data; performing complex conjugate point multiplication on the frequency domain RACH received data and the local frequency domain root sequence, and performing IDFT transform to obtain the multiple RACHs A sequence of time domain correlation values. According to another aspect of the present invention, a detection apparatus based on a random access procedure is provided, the apparatus comprising:

a receiving unit, configured to receive a RACH signal that passes through a wireless channel;

a first processing unit, configured to perform time domain correlation processing on the RACH signal received from the receiving unit and a local root sequence, to obtain a plurality of RACH time domain correlation value sequences;

a setting unit, configured to determine, for each RACH time domain correlation value sequence received from the first processing unit, a corresponding noise power estimation mean Ρ or a maximum peak value pr ^k in the correlation value sequence, and any one of The value calculates a corresponding first detection threshold Th;

a first determining unit, configured to determine, for each search window of the RACH signal time domain correlation value sequence, whether a peak of the RACH time domain correlation value is greater than a corresponding first detection threshold determined by the setting unit, if the determination result is Yes, the peak and its position are recorded; a second determining unit, configured to search for a sequence of time domain correlation values for each RACH signal, wherein there are one or more sets of consecutive plurality of search windows having greater than corresponding Determining whether each peak of the plurality of search windows is greater than a corresponding second detection threshold Th1; wherein ThX _k , _{∞ ί} are among the peaks of the plurality of search windows The maximum value, 0< r < l , / = 1, A, L, and L represents the number of root sequences, k = \, , K , and is the number of groups of the consecutive plurality of search windows;

The second processing unit is configured to convert, for each RACH signal time domain correlation value sequence, a timing position corresponding to a peak value of the second detection threshold determined by the second determining unit into a timing adjustment amount.

The setting unit calculates a first detection threshold rh by using a noise power estimation mean ;; the: the setting unit is based on! ^^ . Γ Calculate the first detection threshold; where, ' = 1, AL , and L denotes the number of root sequences, which is a noise threshold factor greater than 1. The setting unit calculates a first detection threshold by using a maximum peak pr ^k in the sequence of correlation values:

Where is the peak threshold, which ranges from 0<<1.

The setting unit includes: a search window setting module, configured to be based on a sequence length

W

Ncs divides each RACH time domain correlation value sequence into N search windows; where NCS is a preset window length parameter;

a temporary temporary mean calculation module, configured to search for a peak of a RACH time domain correlation value in each search window for each RACH time domain correlation value sequence, and obtain an average value of a RACH time domain correlation value that is less than a times the peak value, Obtain the temporary mean value of noise in the search window 3⁄4 ^iss ; where = Ι, Λ L , and L represents the number of root sequences, j = l, A, N, and N is the number of search windows, 0 < « <1;

The noise power estimation mean calculation module is configured to, for each RACH time domain correlation value sequence, average the noise average values obtained in all the search windows to obtain a corresponding noise power estimation mean value; wherein, 1, 1, A, L, and L represents the number of root sequences;

a first detection threshold setting module, configured to determine a corresponding first detection threshold for each RACH time domain correlation value sequence, where ¹ = ■ Ρ ^ε , a factor greater than 1, '· = Ι, Λ L And L represents the number of root sequences.

The first detection threshold setting module is further configured to determine the value according to a predetermined false alarm probability and a chi-square distribution characteristic.

The second determining unit includes:

a search module, configured to traverse all of the search windows for each RACH time domain correlation value sequence, when there are one or more sets of consecutive multiple search windows having a peak greater than a corresponding first detection threshold τκ\ Recording the peak and its location;

a second detection threshold setting module, configured to determine a corresponding second detection threshold by using a peak value recorded by the search module for each RACH time domain correlation value sequence;

, W is the maximum value among the peaks of the plurality of search windows, 0 ≤ ≤ 1, = 1, A, L, and L represents the number of root sequences, k = A, K, and ^: is a continuous plurality The number of groups of search windows;

The determining module is configured to determine whether the recorded peak value is greater than a corresponding second detection threshold Th determined by the second detection threshold setting module.

The first processing unit includes:

a downsampling processing module, configured to perform a downsampling process on the received RACH signal, and a frequency domain data determining module, configured to perform DFT transform on the data after the downsampling process received by the downsampling processing module, to obtain a frequency domain RACH receiving data. ;

a time domain correlation value determining module, configured to perform complex conjugate point multiplication on the frequency domain RACH received data received from the frequency domain data determining module and the local frequency domain root sequence, and then obtain the plurality of RACH time domain correlation value sequence.

The detection method and detection apparatus based on the random access procedure according to the present invention have the following advantages:

For the calculation of the noise power mean value estimation, the calculation method of calculating the mean value by using the window division can effectively avoid the deviation of the noise power calculation due to the difference of the respective signal-to-noise ratios when multiple users simultaneously select different cyclic shifts of the same sequence. The problem;

Using the distribution characteristics of the noise power, the exact value of the detection threshold is obtained by theoretical calculation, which fully satisfies the requirements of the protocol for the false alarm probability of random access;

By adding the second detection threshold in the window that continuously satisfies the first detection threshold, the problem of the false alarm probability caused by the use of oversampling in the case of high SNR can be effectively avoided, and the false alarm of the preamble by the system is low. Probability. DRAWINGS

1 is a schematic flowchart of a method for detecting a random access procedure according to the present invention;

2 is a block diagram of a baseband processing module of a RACH signal transmitting end specified by the LTE protocol; FIG. 3 is a flowchart of a method for detecting a random access procedure of an LTE system according to an embodiment of the present invention; FIG.

4 is a structural block diagram of a random access based detecting apparatus according to the present invention;

FIG. 5 is a block diagram showing a preferred configuration of a random access procedure based detecting apparatus for an LTE wireless communication system according to another embodiment of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT(S) The preferred embodiments of the present invention are described with reference to the accompanying drawings, and the preferred embodiments of the present invention are intended to illustrate and explain the invention.

The present invention provides a detection method based on a random access procedure. FIG. 1 is a schematic flowchart of a detection method based on a random access procedure according to the present invention. As shown in FIG. 1, the detection method includes the following steps:

Step S102: Perform time domain correlation processing on the received RACH signal and the local root sequence to obtain multiple RACH time domain correlation value sequences.

Specifically, performing time domain correlation processing on the received RACH signal and the local root sequence, and obtaining multiple RACH time domain correlation value sequences generally includes: performing downsampling processing on the received RACH signal; performing data on the downsampled processed data The DFT transform obtains the frequency domain RACH receiving data. The frequency domain RACH receiving data is complex conjugate point multiplied with the local frequency domain root sequence, and then subjected to IDFT transform to obtain the plurality of RACH time domain correlation value sequences.

Step S104: Determine a noise power estimation mean Γ' or a maximum peak value in a sequence of correlation values corresponding to each RACH time domain correlation value sequence, and calculate a corresponding first detection threshold by any one of the values! Where Τ = β · Ρ is a factor greater than 1, = 1, AL, and L represents the number of root sequences. Here, the first detection threshold Γ/ζ is calculated using the noise power estimation mean r' ^se , where ¹ is:

73⁄4 = · /Γ' ^5β calculates the first detection threshold; where , = = AL, and L represents the number of root sequences, β is the noise threshold factor greater than 1, the value is determined by the predetermined false alarm probability and chi-square distribution Characteristics to determine.

The first detection threshold Th is calculated by using the maximum peak value in the correlation value sequence; the first detection threshold is calculated according to 73⁄4 = /^4; wherein, the peak threshold is in the range of 0< <1. Using the noise power estimation mean to calculate the first detection threshold Th; in the case of determining the noise sequence length

The power estimation mean/T' ^Si generally includes: dividing the RACH time domain correlation value sequence into N search windows according to w Ncs; wherein ^ cs is a preset window length parameter; finding the RACH time domain in each search window The peak value of the correlation value, the mean value of the RACH time domain correlation value smaller than the peak value is obtained, and the temporary mean value of the noise in the search window is obtained; wherein, = 1, AL, and L represents the number of root sequences, j = l , A, N, and N is the number of search windows, 0<«<1; and then average the noise average values obtained in all search windows of the RACH time domain correlation value sequence to obtain the mean value of the noise power estimation, = 1, A, L, and L represents the number of root sequences.

Step S106, for each search window of the RACH time domain correlation value sequence, if the peak value in the RACH time domain correlation value is greater than the corresponding first detection threshold Th, record the peak and its position, and perform step 108; otherwise, The process ends.

Step S108, for each search window of the sequence of RACH time domain correlation values, when there are one or more sets of consecutive multiple search windows having peaks greater than the corresponding first detection threshold, determining peaks of the plurality of search windows Whether it is greater than the corresponding second detection threshold 7\ ² , _; , if yes, execute step no; otherwise, the process ends. Where Th = _y .R _p ^k _eak , . ₄ is the maximum value among the peaks of the plurality of search windows, 0 < r < l, i = l, A, L, and L represents the number of root sequences, k = l, J, K, and is continuous The number of groups of search windows.

Step S110: For a peak value larger than the corresponding second detection threshold, convert the corresponding timing position into a timing adjustment amount. 2 is a block diagram of a baseband processing module of a RACH signal transmitting end specified by the LTE protocol. As shown in FIG. 2, the transmitting end first generates a preamble sequence of user information Preamble (protocol defined as ZC sequence 歹 ll ), and then passes subcarrier mapping, and passes through The IDFT completes the OFDM modulation, and adds the cyclic prefix CP to complete the baseband processing of the transmitting end.

In the LTE standard, it is specified that the RACH signal occupies a bandwidth of 1.08 MHz in the frequency domain, and each RACH subcarrier spacing is 1.25 kHz or 7.5 kHz, and there are 864 or 144 RACH subcarriers in the frequency band, of which 839 or The RACH data is placed on 139 subcarriers. These subcarriers are included in the band. The time domain format of the RACH signal is also specified in the standard: the cyclic prefix CP is followed by the preamble of the RACH, and the preamble of the RACH is followed by the guard interval. No signal is sent during the guard interval. The detection sequence of the RACH signal in the LTE wireless communication system according to the detection method of the present invention will be described below using a preamble sequence signal having a format 1, ZC sequence length of 839 and a window length of 13 as a specific embodiment.

FIG. 3 is a flowchart of a method for detecting a random access procedure in an LTE system according to an embodiment of the present invention. It should be noted that, in calculating a first detection threshold, the preferred use of noise power and The corresponding noise threshold factor calculation method is taken as an example. As shown in Figure 3, the detection method includes the following steps:

Step S301: The base station extracts the RACH signal part of the received signal by using the receiving module, performs the downsampling after removing the CP in the time domain, and obtains the frequency domain RACH signal by using the DFT transform.

Step S302, traversing all local frequency domain root sequences allocated to the current cell, performing complex conjugate point multiplication with the local frequency domain root sequence in turn, and then transforming the frequency domain RACH signal into the time domain by IDFT processing. Obtain a corresponding sequence of multiple RACH time domain correlation values, where the RACH time domain correlation value is actually a power value.

Step S303: Determine, for each RACH time domain correlation value sequence obtained in step S302, a corresponding noise power estimation mean value and a corresponding first detection threshold. In this embodiment, for each RACH time domain correlation value sequence, the algorithm for estimating the mean noise power is specifically:

First, the RACH time domain correlation value sequence is divided into N search windows, where, specifically,

839

In the present embodiment, the search window length is 13, and the number of search windows is N 64;

13

The peak value of the time domain correlation value is averaged for the correlation value of ( 0 ≤ « ≤ 1 ) times smaller than the peak value, and the temporary mean value 噪声 of the noise in the window is obtained, where i = lA L ; j = l, A, N ; L represents the number of root sequences.

Next, the noise average value obtained by all the search windows in the RACH time domain correlation value sequence is further averaged to obtain the corresponding noise power estimation mean: = l, L, L.

In this embodiment, the first detection threshold method is calculated by using only the preferred noise power and its corresponding noise threshold factor as an example. For each RACH time domain correlation value sequence, the algorithm corresponding to the first detection threshold is specifically:

When the condition such as the false alarm probability is 10 - ³ is satisfied, the first predetermined threshold value is theoretically calculated according to the distribution of the noise power obeying the chi-square distribution; ^ ( β > 1 ), so that the corresponding first detection Threshold Λ ¹ is:

ΤΗ} = β · . '' ^se , specifically, in this embodiment, y5 = 5 dB.

Of course, similar to the prior art, it is also possible to satisfy the empirical value within the above range. Step S304: For each RACH time domain correlation value sequence, traverse all of the search windows, and detect the peak value of the correlation value in each search window. When the detected result is greater than the first detection threshold, the record satisfies the first detection threshold. Correlation peak and its location; otherwise, there is no RACH signal available in the current window. Step S305: Determine, for each RACH time domain correlation value sequence, a timing position of the transmission preamble starting point according to a preset judgment criterion. Specifically, when the judgment criterion is met, the system determines that the RACH signal and the timing position are found; otherwise, the search window is determined. No RACH signal is available.

In this step, the preset criteria are:

For all search windows of each RACH time domain correlation value sequence, when there are one or more sets of consecutive multiple search windows having peaks exceeding the corresponding first detection threshold, selecting peaks in the consecutive multiple windows The maximum value _k , the r times of the maximum value as the second detection threshold Th^:

TKW

Where U K , K represents the number of groups of such consecutive multiple windows in the search window in the sequence of the i-th RACH time-domain correlation value, which is the second predetermined threshold value and 0 ≤ ≤1.

For the sake of clarity, the described "there are one or more sets of consecutive multiple search windows", for example, a certain RACH time domain correlation value sequence contains 10 search windows, wherein the first and second search windows The peak values are all greater than the corresponding first detection threshold, and the peaks of the fifth, sixth and seventh search windows are also greater than the first detection threshold and the peaks of the remaining five search windows are smaller than the first detection threshold, then There are two consecutive consecutive search windows in the 10 search windows of the RACH time domain correlation value sequence, that is, at this time = 2.

Next, it is determined whether the correlation peak exceeding the corresponding first detection threshold in all K groups exceeds the corresponding second detection threshold, and the correlation peak exceeding the second detection threshold and its timing position are recorded; otherwise, it is considered that the current window is not available. RACH signal.

Step S306, converting the timing position corresponding to the peak value of the detection condition (the above-mentioned determination criterion) in step S305 into a timing adjustment amount, and providing uplink synchronization time information for the user who transmits the preamble.

Step S307, the RACH preamble detection process of the current cell is completed.

The present invention also provides a random access based detection device. FIG. 4 is a structural block diagram of a random access based detection device according to the present invention. As shown in FIG. 4, the device includes: The receiving unit 10 is configured to receive a RACH signal that passes through a wireless channel. The first processing unit 20 is connected to the receiving unit 10, and is configured to receive from the receiving unit 10.

The RACH signal is processed in time domain with the local root sequence to obtain a plurality of RACH time domain correlation value sequences;

The setting unit 30 is connected to the first processing unit 20, and is configured to determine, for each RACH time domain correlation value sequence from the first processing unit 20, a corresponding noise power estimation mean f or a maximum peak value pr ^k in the correlation value sequence. And calculating a corresponding first detection threshold Th from any of the values; wherein, 73⁄4; = ·/Γ^, is a factor greater than 1, = 1, Λ, and L represents the number of root sequences;

The first determining unit 40 is connected to the setting unit 30, and for each search window of the RACH signal time domain correlation value sequence, if the peak value in the RACH time domain correlation value is greater than the corresponding first detection threshold obtained in the setting unit 30 Th , then record the peak and its position;

The second determining unit 50 is connected to the first determining unit 40 for processing the peaks from the first determining unit 40, specifically, for all the search windows of the sequence of time domain correlation values for each RACH signal, when there is a group Or determining whether each of the plurality of search windows is greater than a corresponding second detection threshold Th _k ² _J when the plurality of consecutive plurality of search windows have peaks greater than the corresponding first detection threshold, wherein

_∞i is the maximum value among the peaks of the plurality of search windows, 0≤, ≤1, i = l, A, L, and L represents the number of root sequences, k \Λ , Κ , and is a continuous plurality of The number of groups of search windows;

The second processing unit 60 is connected to the second determining unit 50, and is configured to convert the timing position corresponding to the peak corresponding to the second detection threshold determined by the second determining unit 50 into a timing adjustment amount.

The setting unit calculates the first detection threshold Th by using the noise power estimation mean T ^se ;

Calculating a first detection threshold; wherein, = 1, Λ, and L represents the number of root sequences, which is a noise threshold factor greater than one. The setting unit calculates the first detection threshold Th by using the maximum peak pr ^k in the correlation value sequence: the setting unit calculates a first detection threshold according to ·, wherein, is a peak threshold, and the value ranges from 0 << 1.

The detection sequence of the RACH signal of the detecting apparatus in the LTE wireless communication system according to the embodiment of the present invention is described below with reference to a preamble sequence signal having a format 1 and a ZC sequence length of 839 and a detection window length of 13.

FIG. 5 is a block diagram of a preferred structure of a random access procedure-based detecting apparatus for an LTE wireless communication system according to another embodiment of the present invention. As shown in FIG. 5, the structural block diagram is substantially the same as the structural block diagram shown in FIG. the difference lies in,

The first processing unit 20 includes:

The downsampling processing module 22 performs the downsampling process on the received RACH signal after removing the CP in the time domain;

The frequency domain data determining module 24 is connected to the downsampling processing module 22, and performs DFT transform on the downsampled data in the downsampling processing module 22 to obtain frequency domain RACH receiving data. The time domain correlation value determining module 26 is connected to The frequency domain data determining module 24 traverses all local frequency domain root sequences allocated to the current cell, and performs complex conjugate point multiplication with the frequency domain RACH received signals from the frequency domain data determining module 24 in sequence with each local frequency domain root sequence. Then, the frequency domain RACH signal is transformed into the time domain by IDFT processing, and a corresponding plurality of RACH time domain correlation value sequences are obtained. The RACH time domain correlation value described herein is actually a power value.

The setting unit 30 includes:

The search window setting module 32 is configured to divide each RACH time domain correlation value root sequence into N search windows, wherein, specifically, in the embodiment, the window length is 13, and the number of search windows is N = ― = 64; The noise temporary mean calculation module 34 is connected to the search setting module 32 for searching for the time domain correlation in each search window in the search window for each RACH signal time domain correlation value sequence. The peak value of the value is averaged for the correlation value of "( 0 ≤ « ≤ 1 ) times smaller than the peak value, and the temporary mean value 噪声 of the noise in the window is obtained, where = 1, AL; j = l, A, N; L Representing the number of root sequences; the noise power estimation mean calculation module 36 is connected to the noise temporary mean calculation module 34, and for each RACH signal time domain correlation value sequence, the noise average value obtained by all the search windows is averaged again to obtain the corresponding The estimated noise power is: L;

The first detection threshold setting module 38 is connected to the noise power estimation mean calculation module 36, and when the condition such as the false alarm probability is It) ^-3 is satisfied, the first is theoretically calculated according to the distribution of the noise power obeying the chi-square distribution. The predetermined threshold value (β>\ ), so that for each RACH signal time domain correlation value sequence, the corresponding first detection threshold Th is: Th = β ' P ^e .

The second determining unit 50 includes:

The search module 52 traverses each search window for all search windows of each RACH time domain correlation value sequence, and records these when there are one or more sets of consecutive multiple search windows having peaks greater than the first detection threshold. Peak and its location;

The second detection threshold setting module 54 is connected to the search module 52, and for each RACH signal time domain correlation value sequence, uses the recorded peak value to determine a corresponding second detection threshold 7\ ² , _; , where Tl - _r -R _p ^k _eak , _∞λ is the maximum value among the peaks of the plurality of search windows, o ≤ y, = 1, A, L, and L represents the number of root sequences, k = , A, K , and is continuous The number of groups of multiple search windows;

The determining module 56 is connected to the second detection threshold setting module 54 to determine, for each RACH signal time domain correlation value sequence, whether the recorded peak value is greater than a corresponding second detection threshold 73⁄4 determined by the second detection threshold setting module 54. And recording a point corresponding to a peak larger than the second detection threshold 73⁄4.

Compared with the prior art random detection method and apparatus, the detecting party according to an embodiment of the present invention The method of calculating the mean value of noise power estimation by dividing the window effectively avoids the problem that the noise power calculation is easily deviated in the case of multi-users;

Using the distribution characteristics of the noise power, the exact value of the first detection threshold is obtained through theoretical calculation, which fully satisfies the requirements of the protocol for the false alarm probability of random access;

By setting the second detection threshold, the problem of the false alarm probability in the existing detection method in the case of high signal to noise ratio is solved.

As explained above, the random access procedure based detection method according to the present invention improves the prior art in various aspects. Those skilled in the art should understand that according to

Actual and specific requirements, these aspects can be applied independently or in combination.

The above is only the preferred embodiment of the present invention, and is not intended to limit the present invention, and various modifications and changes can be made to the present invention. Any modifications, equivalent substitutions, improvements, etc. within the spirit and scope of the invention are intended to be included within the scope of the invention.

Claims

Claim

A method for detecting a random access procedure, the method comprising: performing time domain correlation processing on a received random access channel RACH signal and a local root sequence to obtain a plurality of RACH time domain correlation value sequences;

Determining a noise power estimation mean ^rse or a maximum peak value ^se * in the sequence of correlation values corresponding to each RACH time domain correlation value sequence, and calculating by the noise power estimation mean/Γ' or the maximum peak value in the correlation value sequence Corresponding first detection threshold Th;

For each search window of the sequence of RACH time domain correlation values, when there are one or more sets of consecutive multiple search windows having peaks greater than the corresponding first detection threshold, determining peaks of the plurality of search windows Is it greater than a corresponding second detection threshold? 3⁄4, where hlw R , W is the maximum of each of the plurality of search windows, 0 ≤ ≤ l, = 1, A, L, and L represents the root sequence The number, k = A, K, and ^: is the number of groups of the consecutive plurality of search windows;

For peaks larger than the corresponding second detection threshold, the corresponding timing position is converted into a timing adjustment amount.

2. The method according to claim 1, wherein the first detection threshold Th is calculated from the noise power estimation mean / ;; wherein: the first detection threshold is calculated according to = · τ ^{3⁄4 ε} ; wherein, i = l, A , and L represents the number of root sequences, which is a noise threshold factor greater than one.

The method according to claim 1, wherein the first detection threshold ΤΛ ¹ is calculated from a maximum peak pr ^k in the sequence of correlation values: calculating a first detection threshold according to Th = λ · pr ^k ; , is the peak threshold, and its value ranges from 0 << 1.

4. The method according to claim 2, wherein the determining the noise power estimate mean Γ comprises:

According to the length of the W sequence

Wcs is a preset window length parameter;

Find the peak value of the RACH time domain correlation value in each search window, and find the average value of the RACH time domain correlation value smaller than the peak value, and obtain the temporary average value of the noise in the search window, where = 1, AL, and L Represents the number of root sequences, j = l, A, N, and N is the number of search windows, 0 < « <1; and finds the mean value of the noise averages obtained in all search windows of the RACH time domain correlation value sequence, The noise power estimate mean T' ^{Si is obtained} ; where 1, A, L, and L represent the number of root sequences.

The method according to claim 2, wherein the value is determined by a predetermined false alarm probability and a chi-square distribution characteristic.

The method according to claim 1, wherein the performing the time domain correlation processing on the received RACH signal and the local root sequence to obtain a plurality of RACH time domain correlation value sequences includes:

Downsampling the received RACH signal;

Performing DFT transform on the downsampled data to obtain frequency domain RACH received data; performing complex conjugate point multiplication on the frequency domain RACH received data and the local frequency domain root sequence, and performing IDFT transform to obtain the multiple RACHs A sequence of time domain correlation values.

A detecting device based on a random access process, the device comprising: a receiving unit, configured to receive a RACH signal that passes through a wireless channel;

a setting unit, configured to determine, for each RACH time domain correlation value sequence received from the first processing unit, a corresponding noise power estimation mean pr ^se or a maximum peak value in a sequence of correlation values Pr ^k , and calculating a corresponding first detection threshold from the noise power estimation mean p ^e or the maximum peak pr ^k in the correlation value sequence

a first determining unit, configured to determine, for each search window of the RACH signal time domain correlation value sequence, whether a peak of the RACH time domain correlation value is greater than a corresponding first detection threshold determined by the setting unit, if the determination result is Yes, then record the peak and its position;

a second determining unit, configured to search for a sequence of time domain correlation values for each RACH signal, when there are one or more sets of consecutive plurality of search windows having peaks greater than a corresponding first detection threshold Th] Determining whether each peak of the plurality of search windows is greater than a corresponding second detection threshold Th1; wherein, Tli'wR is a maximum value among the peaks of the plurality of search windows,

0<r<l , = 1, A , L, and L represents the number of root sequences, k = l, , K , and is the number of groups of the consecutive plurality of search windows;

The second processing unit is configured to convert, for each RACH signal time domain correlation value sequence, a timing position corresponding to a peak value of the second detection threshold h determined by the second determining unit to a timing adjustment amount.

The device according to claim 7, wherein the setting unit calculates the first detection threshold h by the noise power estimation mean p ^ise : the setting unit calculates the first detection threshold according to ΊΗ=β';; Where i = m, and L represents the number of root sequences, which is a noise threshold factor greater than one.

9. The apparatus according to claim 7, wherein the setting unit calculates a first detection threshold from a maximum peak pr ^k in a sequence of correlation values: the setting unit is based on 73⁄4

Where is the peak threshold, which ranges from 0<<1.

10. The apparatus according to claim 8, wherein the setting unit comprises: a search window setting module, configured to use a sequence length

W

Ncs divides each RACH time domain correlation value sequence into N search windows; where Ncs is a preset window length parameter; a temporary temporary mean calculation module, configured to search for a peak of a RACH time domain correlation value in each search window for each RACH time domain correlation value sequence, and obtain an average value of a RACH time domain correlation value that is less than a times the peak value, Obtaining a temporary mean value of noise in the search window; wherein, = l, AL, and L represents the number of root sequences, j = l, A, N, and N is the number of search windows, 0 < α <1;

The noise power estimation mean calculation module is configured to, for each RACH time domain correlation value sequence, average the noise average values obtained in all the search windows to obtain a corresponding mean value of the noise power, where 1, 1, A, L, And L represents the number of root sequences;

a first detection threshold setting module, configured to determine a corresponding first detection threshold for each RACH time domain correlation value sequence

And L represents the number of root sequences.

11. Apparatus according to claim 8 wherein:

12. The apparatus according to claim 7, wherein the second determining unit comprises:

a search module, configured to traverse all of the search windows for each RACH time domain correlation value sequence, when there are one or more sets of consecutive multiple search windows having peaks greater than a corresponding first detection threshold Th] Recording the peak and its location;

a second detection threshold setting module, configured to determine, by using a peak value recorded by the search module, a corresponding second detection gate K for each RACH time domain correlation value sequence; wherein, Th _k ² .= _r -R _p ^k _eak , W is the maximum value among the peaks of the plurality of search windows, 0 ≤ y ≤ l, i = l, A, L, and L represents the number of root sequences, k = lA, K, and is a continuous plurality The number of groups of search windows;

The determining module is configured to determine whether the recorded peak value is greater than a corresponding second detection threshold 7\ ² determined by the second detection threshold setting module.

13. The apparatus according to claim 7, wherein the first processing unit comprises:

a downsampling processing module, configured to perform a downsampling process on the received RACH signal; a frequency domain data determining module, configured to perform DFT transform on the data after the downsampling process received by the self-dropping sample processing module, to obtain a frequency domain RACH receiving Data